Recent methods for boundary or edge detection built on Deep Convolutional Neural Networks (CNNs) typically suffer from the issue of predicted edges being thick and need post-processing to obtain crisp boundaries. Highly imbalanced categories of boundary versus background in training data is one of main reasons for the above problem. In this work, the aim is to make CNNs produce sharp boundaries without post-processing. We introduce a novel loss for boundary detection, which is very effective for classifying imbalanced data and allows CNNs to produce crisp boundaries. Moreover, we propose an end-to-end network which adopts the bottom-up/top-down architecture to tackle the task. The proposed network effectively leverages hierarchical features and produces pixel-accurate boundary mask, which is critical to reconstruct the edge map. Our experiments illustrate that directly making crisp prediction not only promotes the visual results of CNNs, but also achieves better results against the state-of-the-art on the BSDS500 dataset (ODS F-score of .815) and the NYU Depth dataset (ODS F-score of .762).
An Aspergillus versicolor isolated from sediment collected from the Bohai Sea, China, yielded the new dimeric diketopiperazine brevianamide S (1), together with three new monomeric cometabolites, brevianamides T (2), U (3), and V (4). Structures were determined by detailed spectroscopic analysis. Brevianamide S exhibited selective antibacterial activity against Bacille Calmette-Guérin (BCG), suggestive of a new mechanism of action that could inform the development of next-generation antitubercular drugs.
Selective removal of organic pollutants
by advanced oxidation methods
has been receiving increasing attention for environmental remediation.
In this study, a novel catalyst, which can selectively oxidize phenolic
compounds (PCs) based on their hydrophobicity, composed of metal–organic-framework-derived
Fe/Fe3O4 and three-dimensional reduced graphene
oxide (rGOF) is designed for peroxydisulfate (PDS) activation. This
heterogeneous PDS activation system can completely degrade hydrophobic
PCs within 30 min. By investigating the hydrophobic properties of
eight representative PCs, a positive correlation between the hydrophobicity
of PC and the reaction kinetics is reported for the first time. The
selective removal stems from the strong interaction between highly
hydrophobic PCs and the catalyst. Moreover, the mechanism investigation
shows that the degradation reaction is triggered by interface reactive
oxygen species (ROS). Our study reveals that the selective degradation
of organic pollutants by PDS activation depends on the hydrophilic
and hydrophobic properties of the pollutant and catalyst. The reported
results provide new insights into a highly selective and efficient
PDS activation system for organic pollutant removal.
Tripterygium wilfordii Hook F. (TwHF) based therapy has been proved as effective in
treating rheumatoid arthritis (RA), yet the predictors to its response remains unclear. A
two-stage trial was designed to identify and verify the baseline symptomatic predictors of
this therapy. 167 patients with active RA were enrolled with a 24-week TwHF based therapy
treatment and the symptomatic predictors were identified in an open trial; then in a
randomized clinical trial (RCT) for verification, 218 RA patients were enrolled and
classified into predictor positive (P+) and predictor negative (P−) group, and were randomly
assigned to accept the TwHF based therapy and Methotrexate and Sulfasalazine combination
therapy (M&S) for 24 weeks, respectively. Five predictors were identified (diuresis,
excessive sweating, night sweats for positive; and yellow tongue-coating, thermalgia in the
joints for negative). In the RCT, The ACR 20 responses were 82.61% in TwHF/P+ group,
significantly higher than that in TwHF/P− group (P = 0.0001) and in M&S/P+ group
(P < 0.05), but not higher than in M&S/P− group. Similar results were
yielded in ACR 50 yet not in ACR 70 response. No significant differences were detected in
safety profiles among groups. The identified predictors enable the TwHF based therapy more
efficiently in treating RA subpopulations.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.